The determination of uranium in low grade ores, some industrial raw materials (e.g. phosphates) and industrial wastes (e.g. coal ashes) at concentrations below 200 ppm Is very important for various reasons, and poses a serious analytical problem. In these materials uranium is often accompanied by elements emitting K lines which interfere with the most intensive U Lα and U Lβ1 lines. The most serious interferences for U Lα lines are from Sr and Rb sometimes present in analysed materials in comparable or excessive quantities. Also Mo, Zr and Y can interfere.

The properties of particle-induced X-ray emission, PIXE, and secondary target mode X-ray fluorescence, XRF, applied to the analysis of unprepared drill cores in open air have been evaluated. Typical detection limits for elements heavier than Mg have been determined for a PIXE-system with an external 2.55 MeV proton beam and for an XRF-system with Ti, Mo and Tb secondary targets. These two systems were found to have similar detection limits for most elements in a typical geological sample. The heterogeneous composition of drill cores prevents the performance of accurate matrix corrections, though calculations using fundamental parameters show that in the PIXE analysis of elements heavier than Ca, these corrections are much less sensitive to variations in the matrix composition than in the XRF analysis.

X-ray fluorescence method using Compton scattering for the determination of ash in coal has been systematically studied for application to Indian Coal. Fluorescent intensities from major constituents such as Al2O3, SiO2, TiO2, Fe2O3, CaO and their various combinations were used in conjunction with the reciprocal of the Cr Kα Compton intensity in a regression equation. Varying degrees of correlations were obtained between the values of ash determined by conventional methods and those calculated from X-ray data. It is found that Ti fluorescent intensity plays a major role in the regression equation and its contribution cannot be ignored if the Cr X-ray tube is used.

If we want to evaluate the present state of X.R.S., we should do this in perspective with other methods, e.g. Optical Emission and Atomic Absorption Spectrometry.

One way to compare would be to estimate the number of papers published or given at conferences on these three methods. Table 1 gives a survey af some meetings. It is evident that X.R.S. is less talked about than the other methods. The rejuvenation of OES, thanks to the Plasma sources, can also clearly be seen.

The origin of the background under the signal in the wavelength dispersive XRF has been investigated. The methods of minimizing the background are presented. The bremsstrahlung is eliminated by using a primary filter in front of the X-ray tube. The background generated by the X-ray detector is decreased with a pulse risetime and a pulse interval analyzer. For example, in the fixed uranium spectrometer channel a 500-fold improvement of the peak-to-background ratio was achieved.

Continued development of mercuric iodide (HgI2) detectors for x-ray spectroscopy at room-temperature has led to a considerable improvement in energy resolution and a better understanding of the various detector parameters which affect sensitivity. The basic properties of a mercuric iodide detector and some of its characteristics pertinent to x-ray fluorescence analysis have been previously reported (1,2,3). In this paper we present results of studies to determine the shape o£ peaks and continuum background. Also, the use of Hgl2 in characterizing water pollutants by XRF analysis has been investigated and compared to cryogenically cooled Si(Li) and room-temperature proportional counter systems.

In order to improve the overall performance of energy dispersive x-ray secondary emission spectrometry one can make use of polarized x-rays. We have used polarized x-rays produced by 90° scattering to reduce the background intensity due to the primary x-ray source.

An effort has been made to optimize the polarizing scatterer for the elements being analyzed. A discrepancy appears in the theory of such optimization between a simple one point formula and a more elaborate integral formula. Possible reasons for such disparities might lie in 1) collimator geometric effects, 2) multiple scattering and 3) actual primary intensity. The first problem has been dealt with in an unpublished manuscript. This report concerns itself with point 2, multiple scatter, and in particular with double scatter including polarization effects.

Using low-energy electron induced X-ray spectroscopy (LEEIXS) with a cold cathode tube as excitation source and appropriate calibrations, quantitative surface composition informations are obtained. The results concern anodic film thicknesses and carbon impurities, In addition, qualitative informations on the chemical state of the film components are deduced from the fine structure of X-ray emission bands. The results concern oxygen K emission spectra from CuO and Cu2O, SnO, SnO2 and anodized Sn. It is also shown the capabilities of LEEIXS for gaining element-depth profiles, One example pertaining to phosphorus impurities in Al2O3 anodic films is given. Finally advantages and disadvantages of LEEIXS over ion-induced X-ray spectroscopy are discussed.

X-ray proportional gas detectors are used in X-ray diffraction and X-ray fluorescence and emission analysis. The amplitude of the electrical pulses produced by these detectors is directly proportional to the energy of the incident X-rays. In most instances, the energy of the incident photons will be utilized completely in producing photo-ionization in the outermost shells of the detector atoms. If the energy of these photons is greater than the absorption edge of the atoms in the detector, a certain fraction of them will produce photo-ionization of an inner shell. This in turn results in the production of a characteristic X-ray. If this characteristic X-ray is absorbed by the detector, then the original X-ray may still be considered to have deposited the whole of its energy within the detector. In this case the amplitude of the detector output pulse will be the same as if the original X-ray photon had been absorbed entirely by photo-ionization of the outer shells. Because of the relatively low absorption of a detector for its own characteristic radiation, there is a significant probability that the detector characteristic X-ray will escape from the detector without producing further photo-ionization.

A device was made for preparing accurate definition of surface, depth and volume of liquid x-ray fluorescence specimens.

The apparatus used in conjunction with a specimen holder and plastic film window material accurately and consistently forms a flat bobble-free analysis window on the open face of the specimen holder. The specimen holder in the form of a shallow cylinderical cup is slightly over filled and covered by the plastic film. Placement of the mating leveling apparatus over the film squeezes out trapped air bubbles, levels the exposed face of the specimen, draws the plastic film tightly over the exposed face of the specimen and allows easy installation of a film retaining O-ring to maintain the specimen material in a level state within the holder.

After the pioneering work of HOFFER et al. on fluorescent scanning of natural iodine in the thyroid, several authors have suggested the use of X-ray fluorescence analysis to follow in vivo the distribution and elimination of stable tracers injected prior to the investigation. The normal method of performing such studies is to inject iodine containing contrast media and then follow the variation in iodine concentration by in vivo measurements over the organ or tissue of interest.

A theoretical treatment of the application of X-ray fluorescence in in vivo studies has been made by TINNEY by means of analytical calculations.

Lead is a non-essential toxic metal of considerable topicality. It is accumulated in the skeleton, which contains 90% of the total body burden.

It is known that during the Roman period the intake of lead was high. The Romans used kitchen utensils made of lead and lead pipes for drinking water. The most important source of lead for people of the upper social classes was sapa, a compound added to wine to sweeten and perserve it.

Using non-destructive X-ray fluorescence analysis, lead and barium concentrations of archaeological Roman bones have been measured. We have earlier used a similar technique to determine lead concentration in the human skeleton in vivo.

The potential of energy-dispersive X-ray fluorescence spectrometry for analysis of refractory metals and WC-based hard metals is investigated. Both, photon excitation by filtered tube radiation and by the characteristic, lines of a secondary target are employed. Both excitation systems give good results. If the counting times are adjusted to account for the lower sensitivity of energy dispersive as opposed to wavelength dispersive X-ray spectrometry the detection limit and precision data are comparable. The multielement analyses of interest in these applications that comprise an energy range of 5 keV or more are better handled by direct excitation with filtered tube radiations than either by secondary target excitation or by wavelength dispersive X-ray spectrometry.

An improved method for sample preparation and determination of minor elements in catalytic materials by radioisotope induced X-ray fluorescence is presented. The sample preparation is simple and rapid, it utilizes both and internal standard procedure as well as a thin film technique. Synthetic standards are prepared simply by mixing thoroughly the appropriate amounts of the elements of interest as their respective oxides with the matrix and the internal standard. The above standards are used to calculate the relative-fluorescence ratio factors (FJL) for each element of interest relative to the internal standard.

Some of the different types of deviations (sample preparation, standard counting, instrumental, etc) will be shown for typical determinations applying experimental methods. Finally the application of this technique for the determination of Cobalt, Nickel, Molybdenum, Platinum and Palladium in catalytic materials is presented.

The Inorganic Analytical Laboratory of Owens-Corning Fiberglas is routinely using a Philips PW1600 x-ray spectrometer for the analysis of glass and glass raw materials. The spectrometer is equipped with 11 fixed channels, two scanning channels, and an energy dispersive detector. Operation of the spectrometer and data compilation is accomplished using a Digital Equipment Corp. PDP11/34 computer. The x-ray is controlled by an RSX-11M operating system, and the Philips “Alphas” software package is used to process data using the Lucas- Tooth Pyne (LP), Lachance-Traill (LT), or Rasberry-Heinrich (RH) correction models.

Residual analyte intensity exists in a spectrometer when, for example, the x-ray source is contaminated with the analyte. A blank specimen, identical to the sample but devoid of the analyte, may be used to determine this residual analyte intensity, and, thereby, allow quantification of that intensity due to analyte in the sample. It is generally considered that a single pure material could serve as a blank for many analytes in many matrices. In practice, at least for low average atomic number matrices such as glasses, the reflectivity of similar samples may differ substantially, rendering inaccurate this simple blank subtraction. In such cases a blank is needed for each sample, where the reflectivity of the blank is matched to that of the sample. In order to eliminate the need for generation of blanks to match each of the samples, a general method to quantify the net residual analyte intensity as a function of specimen reflectivity by measurement of two pure material blanks is proposed herein.

This paper describes a computer method for wavelength dispersive (WD) qualitative X-ray fluorescence (XRF) analysis. It determines the elements, spectral lines, wavelengths, reflection angles and peak intensities of the first and second order reflections in less than half a minute of time using an IBM Series/1 minicomputer. The resolution and precision are significantly better than the energy dispersive (ED) method and when combined with high speed computer recording the speed is comparable.

Inherent in the use of radioisotope sources with secondary fluorescers is the background produced by scattering of the source photons from the exciter system. A Monte Carlo program has been developed that is capable of simulating the backscattered photon spectrum as a function of the system geometry, including shielding and collimation variations. This computer program generates the scattered photon spectrum incident on both the sample and detector. The program is applied to a commercially available exciter system to study the effect of specific geometric design changes on the scattered spectrum.